The goal of this project is to establish predictive algorithms useful for drug discovery and development covering six potassium (K+) channels that are important targets associated with potential cardiotoxicity of drugs. To accomplish this larger goal, we first plan to establish a clonal cell line that heterologously expresses each targeted K+ channel, and to develop ligand binding and functional assays using those cell lines. CHO cells are used as host cells because they are not known to natively express potassium channels. During this Phase I effort, we propose to develop 6 radioligand binding assays and 12 functional assays. We will then screen through each assay a small, focused chemical library with known chemical structures that are expected or known to inhibit potassium channels, thus establishing Structure-Activity Relationships for the six targets. These chemicals will include peptide toxins initially, but we also propose to identify and utilize small molecule inhibitors, such as those found to inhibit the human ether-a-go-go (HERG) potassium channel. We have demonstrated with the HERG channel using a similar approach and methods that we can establish powerful predictive algorithms, based on the compounds' chemical structural features and data generated on inhibition of known HERG inhibitors in the assays. These predictive algorithms can then be applied in silico to new chemical structures or substructures to assess the probability that such compounds will interact with each K+ channel. The end-product of this proposed project may become a powerful tool for researchers in drug discovery to be able to predict potential cardiac toxicity (or antiarrhythmic activity) early in drug development, potentially even on a virtual basis before a compound is synthesized.